Elamur Research⌕
☰
home / papers / 2018 / The Circumstantial Event Ontology (CEO) and ECB+/CEO: an Ontology and Corpus for Implicit Causal Relations between Events

The Circumstantial Event Ontology (CEO) and ECB+/CEO: an Ontology and Corpus for Implicit Causal Relations between Events

Summary

The authors examine causal relations between events that are not explicitly stated in news reports. The reader reconstructs the chain on their own, e.g. "impact — capsizing — sinking — death". For such relations they propose the notion of "circumstantial" relations. An event explains why the following one happened but does not make it necessary. Two types are distinguished: episodic relations between specific events in a text and semantic relations between event classes. The main resource of the work is the CEO ontology for calamity events. It extends ESO: each class describes the state of the participants before, during and after the event. A circumstantial relation between classes is inferred when their assertions concern a change in one and the same property. Classes and roles are mapped to FrameNet, so the ontology can be populated with the output of SRL systems. For evaluation, the authors manually annotated a corpus of episodic circumstantial relations based on ECB+. Simple heuristics yield promising precision, but recall is still low. The value of the work lies in the fact that relations between events rest on shared properties rather than on connectives. Therefore they also work beyond the boundaries of a sentence or paragraph.

Key points

  • CEO is an ontology in OWL2 format, distributed under the CC-BY-SA license via the newsreader/eso-and-ceo repository on GitHub. Classes are manually mapped to FrameNet and SUMO (e.g., ceo:Arson → fn:Action, sumo:Arson).
  • Properties are either binary (linking two roles) or unary (a Boolean value or a relative change "+"/"−"). If a role is not found in the text, a blank node is created via an OWL existential restriction.
  • The order of events is implicitly determined by the position of the shared property in the descriptions of situations. Example: "fire exists" is a post-situation of ceo:Arson, a during-situation of ceo:Fire and a precondition of ceo:ExtinguishingFire.
  • The ECB+/CEO corpus is a manual annotation of episodic circumstantial relations on top of ECB+. When comparing only property predicates without taking order into account, we obtain P = 0.114, R = 0.165, F1 = 0.135. Full assertions give somewhat higher precision but lower recall.
  • In the future, the authors plan to develop a reasoner and test the model on a QA task with "why did the event happen" questions.

The middle of the paper is missing from the pasted text. Therefore the summary does not include the corpus size, inter-annotator agreement scores or the results for full assertions.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics Knowledge graphs & ontologies

Cite

Cite as Roxane Segers, Tommaso Caselli and Piek Vossen. 2018. The Circumstantial Event Ontology (CEO) and ECB+/CEO: an Ontology and Corpus for Implicit Causal Relations between Events. Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018).

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{segers2018circumstantial,
  title = {{The Circumstantial Event Ontology (CEO) and ECB+/CEO: an Ontology and Corpus for Implicit Causal Relations between Events}},
  author = {Roxane Segers, Tommaso Caselli and Piek Vossen},
  booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)},
  year = {2018},
  url = {https://aclanthology.org/L18-1725/}
}

References

17 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Bartalesi Lenzi, V., Moretti, G., and Sprugnoli, R. (2012). Cat: the celct annotation tool. In LREC, pages 333–338. Björkelund, A., Hafdell, L., and Nugues, P. (2009). Multilingual semantic role labeling. In Proceedings of CoNLL-2009, Boulder, CO, USA. find
  2. Brown, S., Bonial, C., Obrst, L., and Palmer, M. (2017). The rich event ontology. In Proceedings of the Events and Stories in the News Workshop, pages 87–97, Vancouver, Canada, August. Association for Computational Linguistics. Caselli, T. and Morante, R. (to appear). Agreements and Disagreements in… find
  3. Chambers, N. and Jurafsky, D. (2010). A database of narrative schemas. In Proceedings of the 9th Language Resources and Evaluation Conference (LREC2010). find
  4. Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and psychological measurement, 20(1):37–46. find
  5. Cybulska, A. and Vossen, P. (2014). Using a sledgehammer to crack a nut? lexical diversity and event coreference resolution. In Proceedings of the 9th Language Resources and Evaluation Conference (LREC2014), Reykjavik, Iceland, May 26-31. our card
  6. De Beaugrande, R. and Dressler, W. (1981). find
  7. Textlinguis-tics. New York: Longman. de Lacalle, M. L., Laparra, E., Aldabe, I., and Rigau, G. (2016). A multilingual predicate matrix. In Proceedings of Language Resources and Evaluation Conference (LREC 2016). find
  8. Fellbaum, C. (1998). WordNet: an electronic lexical database. find
  9. MIT Press. Landis, J. R. and Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, pages 159–174. find
  10. Niles, I. and Pease, A. (2001). Towards a standard upper ontology. In Proceedings of FOIS-Volume 2001. find
  11. ACM. Prasad, R., Miltsakaki, E., Dinesh, N., Lee, A., Joshi, A., Robaldo, L., and Webber, B. L. (2007). The penn discourse treebank 2.0 annotation manual. find
  12. Ruppenhofer, J., Ellsworth, M., Petruck, M., Johnson, C., and Scheffczyk, J. (2006). find
  13. FrameNet II: Extended Theory and Practice. International Computer Science Institute, Berkeley, California. Segers, R., Rospocher, M., Vossen, P., Laparra, E., Rigau, G., and Minard, A. (2016). The event and implied situation ontology: Application and evaluation. In Proceedings of Language Resources… find
  14. Segers, R., Caselli, T., and Vossen, P. (2017). The circumstantial event ontology (ceo). In Proceedings of the Events and Stories in the News Workshop, pages 37–41, Vancouver, Canada, August. Association for Computational Linguistics. find
  15. Van den Akker, C., Aroyo, L., Cybulska, A., van Erp, M., Gorgels, P., Hollink, L., Jager, C., Legene, S., van der Meij, L., Oomen, J., van Ossenbruggen, J., Schreiber, G., Segers, R., Vossen, P., and Wielinga, B., (2010). Historical Event-based Access to Museum Collections, pages 1–9. CEUR-WS (onlin… find
  16. Vossen, P., Caselli, T., and Kontzopoulou, Y. (2015). Storylines for structuring massive streams of news. In Proceedings of the 1st Workshop on Computing News StoryLines (CNewS 2015) at the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conferenc… find
  17. Language Resource References VV.AA. (2017). WordNet. unspecified, ISLRN 379-473-059-273-1. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/segers-2018-circumstantial-event-ontology-ceo-ecb/